ANTIBODY ANALYSIS By High Performance Capillary Electrophoresis

Analysis of Immunoglobulin G

Label free analysis using the HPCE-512

Capillary Electrophoresis (CE) Application Note CE-01 Antibody Analysis– High Performance CE

ABSTRACT Analysis was carried out on several

mammalian serum IgGs and commercially

available Control Standard IgG using

the HPCE-512 instrument to assess IgG

stability based on fragmentation, purity and

heterogeneity. Fragmentation pattern and base-line resolved

non-glycosylated heavy chain peaks were observed, with excellent reproducibility data

obtained (peak time RSD less than 0.03%; quantitation RSD <1%). Accurate quantitation

of the non-glycosylated species was also achieved, which came to ~9.6% of total heavy

chain. This application note demonstrates how high

performance capillary electrophoresis, using a

512 pixel detector, can be used to assess the

fragmentation pattern and post-translational

modifications of IgG. The same strategy may

be applied in assessing the stability and

efficacy of biopharmaceutical mAbs.

INTRODUCTION In order to assess the bioactivity and safety of mAbs (recombinant monoclonal antibodies)

as biopharmaceutical products, it is important

to characterise the molecule and any

impurities present. Not only is the information

crucial for the assessment of the suitability of

a potential biopharmaceutical product, it is

also important for optimisation and QA/QC of

the manufacturing process, storage conditions

and life-time of the drug.

Out of five defined classes of antibody, all

currently licensed mAbs are of the

immunoglobulin G (IgG) class with

applications in the both therapeutic and

immunodiagnostic sectors. Recently, many

research groups have reported both qualitative

and quantitative mAbs analysis by SDS-CGE

(SDS-Capillary Gel Electrophoresis) systems.

Some have reported the detection of fragments

from monoclonal antibody samples, which

correspond to H2L, H2, HL, H and L portions of

the antibodies. The fragmentation was

believed to be caused by incomplete

assembly, lack of intermolecular disulfide

bridges and/or proteolysis due to microbial

contamination. The level of fragmentation has

been used as a measure of antibody stability.

The effects of pH, temperature and buffer

system on fragmentation have also been

studied using SDS-CGE.

MATERIALS AND METHODS Mammalian serum IgGs were prepared in

reducing and non-reducing conditions. Reduced samples were prepared with a

proprietary methodology and incubated for 10 minutes at 95 °C. Non-denaturing samples were

fragmented by heating for 10 minutes at 95°C. The reduced control standard IgG was

prepared with the same proprietary methodology as above and heated for 5

minutes at 95°C.

Separation by capillary electrophoresis was

performed in a commercial run buffer in bare

fused silica capillaries of 50 μm internal

diameter and a separation length of 20 cm.

Total length of capillary was 32 cm. All serum

IgG samples were injected electrokinetically at

5 kV for 10s and separated at 18 kV for 25

minutes. Control standard IgG samples were

pressure injected at 4 psi for 10 seconds

and separated at 18 kV for 25 minutes. The

capillary was held at 22 °C during the

separation. All detection was at 214 nm. At the

beginning of the experiment and between the

runs, the capillary was conditioned using a

proprietary methodology. Between runs the

capillary was rinsed with 1M NaOH for 5

minutes followed by 0.1M NaOH, 0.1M HCl and

run buffer for 5 minutes each. After

conditioning, a buffer conductivity check was

performed.

Analysis of the separations was performed

using the proprietary Equiphase Vertexing

Algorithm (EVA) and Generalised Separation

Transform (GST) algorithm. GST is a

method of combining the data from the 512

pixels in a natural way which preserves the

peak shape information of the

electropherograms while at the same time

maximising the signal-to-noise ratio. Typically

a 10- fold increase in signal-to-noise using

GST is observed compared to single

electropherograms. EVA is an advanced

pattern-recognition tool which maximizes the

system resolution. In EVA the

electropherograms are first analyzed to find

local peaks. These are used first to perform

vertexing (determine the point of origin of the

bands) and then to produce a signal output.

Capillary Electrophoresis (CE) Application Note CE-01 Antibody Analysis– High Performance CE

Figure 3: An overlay of EVA-processed data of 8 consecutive runs of Commercial control standard IgG In reducing conditions.

Table 1:

Mean, Standard Deviation,

Relative Standard Deviation of

peak time and relative

quantitation for Light Chain,

Non-Glycosylated Heavy Chain

and Heavy Chain of Control

Standard IgG separated using

denaturing conditions

Figure 2: Commercial control standard under denaturing conditions, EVA processed data, LC Light

Chain, HC Heavy Chain, NG-HC Non-Glycosylated Heavy Chain.

0.001

0.006

0.002

0.003

0.007

0.005

12 13 15 16 17 TIME (MINS)

ABS (ARB)

LC

NG-HC

HC

0.007

0.001

0.003

0.005

ABS (ARB)

4000 6000 8000 10000 12000

LOW MOLECULAR WEIGHT IMPURITIES

LC

NG-HC

HC

Capillary Electrophoresis (CE) Application Note CE-01 Antibody Analysis– High Performance CE

RESULTS AND DISCUSSION Three mammalian serum IgGs (bovine, goat and rabbit) were separated under reducing

and non-reducing conditions. In reducing conditions, the SDS-protein complex was

heated for 10 minutes at 95°C in the presence of a reducing agent. This treatment breaks the

covalent disulphide bond between the heavy and light chain (HC and LC respectively), reducing the

IgG structures to two species with distinct molecular weights. In non-reducing conditions,

the SDS-protein complex was heated at 100°C for 10 minutes in the absence of a reducing

agent to facilitate the fragmentation process of IgG structure.

Figure 1 shows the EVA processed data of heat-treated, non-reduced bovine serum IgG.

A profile of 8 species with distinct migration

times was observed, where the LC and HC

could be easily identified based on migration

time information derived from the GST profile

of the reduced IgG (data not shown). With the

latest migrating peak assigned as the whole

IgG molecule, the rest of the peaks were

identified as 2HC+1LC, 2HC, 1HC+1LC and

low-level impurities present in the sample.

The same fragmentation pattern was observed

in both non-reduced goat and rabbit serum

IgGs – with entire IgG, 2HC +1LC, 2HC,

1HC+1LC, HC/LC portions and low-level

impurities (data not shown). The migration time

of each portion varies across the three

mammalian IgGs examined.

Given SDS-CGE’s superior resolving power

over traditional SDS-PAGE, it is possible to

examine the heterogeneity of post-translational

modifications in IgG through mass-based

separation. One important aspect of

post-translational modification is glycosylation.

HPCE resolving power in the SDS-CGE system

with HPCE technology can be demonstrated by

the base-line resolved separation of the non-

glycosylated (NG HC) and glycosylated (HC)

heavy chain molecules. The sample used was a

commercial control standard IgG where a known

quantity of non-glycosylated heavy chain is

present. Figure 2 shows the EVA processed data for the commercial control standard IgG,

where the non-glycosylated heavy chain is

baseline-resolved from the glycosylated heavy

chain.

Figure 3 shows the overlay of EVA-processed data of 8 consecutive runs of commercial

control standard IgG in denaturing conditions.

The analysis results of those 8 runs are

summarized in Table 1. The relative standard deviations (%RSD) for migration of all three

species (light chain, non-glycosylated heavy

chain and glycosylated heavy chain) are all

<0.03% while the quantitations of each species

(%LC, %NG-HC, and %HC) are all <1% RSD.

Through the reproducibility study, the precision

of the HPCE instrument in both resolution and

quantitation is clearly demonstrated. Based on

quantitation analysis, the percentage of non

-glycosylated heavy chain present in the

sample was calculated, which came to 9.55% ±

0.007 of the total heavy chain, compared to the

supplier’s figure of 9.5%.

0.005

0.015

0.025

0.035

0.045

ABS (ARB)

12 13 14 15 16 22 21 20 19 18 17 24 23

TIME (MINS)

LC

LOW MW IMPURITIES

HC

1HC+ 1LC

2HC 2HC+ 1LC

WHOLE iGg

Figure 1: EVA processed data of SDS-CGE separation of bovine serum lgG using commercial SDS gel buffer under non-reducing conditions

Capillary Electrophoresis (CE) Application Note CE-01 Antibody Analysis– High Performance CE

Light chain glycosylation can also be

observed using the proprietary protocols. Due

to the presence of a ‘shoulder’ on the light

chain peak of some antibody samples, an

alternative CGE method was utilised in an

attempt to further resolve any potential

glycoforms present in these samples.

The separation length of the capillary was

increased to 40cm in order to improve

resolution between the glycosylated and non-

glycosylated light chains of the antibody

(Figure 4), and some adjustments made to the buffer regime.

Conclusion This application note demonstrates how

SDS-CGE on the HPCE-512 system can be

used to obtain valuable information on lgG

sample stability, purity and heterogeneity in a

rapid and reproducible way. Fragmentation

patterns and base-line resolved

non-glycosylated heavy chain peaks were ob-

served, with excellent reproducibility (peak

migration time <0.03%; quantification <1%

RSD). Accurate quantification of the non-glycosylated species was also achieved , which came to

9.55% of total heavy chain. Light chain glycosylation was also observed. When applied to the production of biopharmaceutical products, the HPCE system can be a powerful tool in the assessment of the bioactivity and safety potential products. This information will be vital in the optimisation and quality control of the manufacturing process, storage conditions and life-time of the biopharmaceutical product.

Figure 4: Glycosylation on the light chain of an antibody by CGE analysis.

EVA

0.0014

0.0012

ABS (ARB)

0.0018

0.0016

0.0004

0.0006

0.0008

0.0002

10 11

12 13 14 TIME (MINS)